INDEX

Explanations

engines, loving, happy, thread, Palace

static images

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 aaye

0.54

 conveyed

0.52

 cataly

0.51

loxy

0.50

 актива

0.49

 postulated

0.49

 пусть

0.49

 interdependent

0.49

 тобто

0.48

 observed

0.47

POSITIVE LOGITS

Culture

0.56

0.53

0.52

0.48

Source

0.48

 Culture

0.47

Qu

0.46

Line

0.46

Activations Density 0.000%